Download as a PDF

Journal of Health Politics, Policy and Law

Why Some Market Reforms Lack Legitimacy in Health Care Miriam Laugesen University of California, Los Angeles

Market-oriented health policy reforms in the 1980s and 1990s generally included five kinds of proposals: increased cost sharing for patients through user fees, the separation of purchaser-provider functions, management reforms of hospitals, provider competition, and vouchers for purchasing health insurance. These policies are partly derived from agency theory and a model of managed competition in health insurance. The essay reviews the course of reform in five countries that had a national health service model in place in the late 1980s: Italy, New Zealand, Spain, Sweden, and the United Kingdom. Special consideration is given to New Zealand, where the market model was extensively adopted but short lived. In New Zealand, surveys and polls are compared to archival records of reformers’ deliberations. Voters saw health care differently from elites, and voters particularly felt that health care was ill suited to commercialization. There are similarities across all five countries in what has been adopted and rejected. Some market reforms are more legitimate than others. Reforms based on resolving principal-agent problems, including purchaser-provider splits and managerial reforms, have been more successful, although cost sharing has not. Competition-based reforms in financing and to a lesser extent in provision have not gained legitimacy. Most voters in these countries see health care as different from other parts of the economy and view managerial reforms differently from policies that try to make health care more like other sectors.

Abstract

The author acknowledges Mark Schlesinger’s help in extending the scope of this essay. The essay draws on research funded by Health Research Council of New Zealand Post-Graduate Scholarship 96–405 and the University of Melbourne Faculty of Arts Fieldwork Fund. Previous versions of this essay were presented at the annual meetings of the American Political Science Association in 2003 and 2004. Thanks to the Ministry of Health for making their files available. Readers may contact the author at [email protected] for the referencing numbers of these files. Journal of Health Politics, Policy and Law, Vol. 30, No. 6, December 2005. Copyright © 2005 by Duke University Press.

Published by Duke University Press


1066


In the late 1980s and early to mid-1990s, governments in a number of countries introduced market-like processes into national health service (NHS) health care systems. In a typical NHS, also called an integrated model (Organisation for Economic Co-operation and Development [OECD] 1992), the government is the single payer and owner of hospitals, with contributions typically made through the tax system rather than through or by employers. These types of reforms are also known as managed competition reforms and internal market reforms and were adopted in Italy, New Zealand, Spain, Sweden, and the United Kingdom in the late 1980s and early 1990s. Market-based approaches usually involved a combination of the following five policies: (1) increased cost sharing for patients through user fees, (2) the separation of purchaserprovider functions, (3) management reforms of hospitals, sometimes paired with profit or return on capital expectations, (4) provider competition, and (5) vouchers for purchasing health insurance. Reforms were based on economic theories and the experience of the governments as they reformed other sectors. Market reforms are interesting, because although governments in many countries found market policies attractive, they were “dangerous to implement” (Light 2001a: 1151). In some cases, these reforms led to subsequent reaction against markets and toward increased government financing and regulation. This is puzzling, since voters are willing to accept freer markets in many areas. They do not reject the consumer role in all areas of public policy. The question explored here is why is market reform so difficult and contentious? To understand why market reform has not been well accepted in health care, the political responses to market reform in NHS systems are explored in five countries (Italy, New Zealand, Spain, Sweden, and the United Kingdom). To answer this question, it is instructive to understand why laissez-faire approaches and competition are tolerated in some areas of public policy more than in health care. Drawing on scholarship in public opinion and public policy, a theory of why market reforms failed in New Zealand is developed based on tests of two hypotheses. First, voters view health care as different. Whereas policy-making elites in the 1990s thought health care needed solutions similar to state-owned businesses, voters tend to regard health care as different from the businesses reformed by government into profitable entities. As a result, they are less willing to support market reform policies of health care. Second, creating health care markets and entrepreneurial behavior is more difficult than making other kinds of markets, for a number of reasons. First, the scale is larger,



Laugesen

■

Market Reforms in NHS Countries 1067

in terms of resources and people, and the task of encouraging formerly bureaucratic organizations to embrace competition is challenging. Second, professionals occupy a different position and are a sympathetic group that can garner support of public opinion. Third, in pursuing the objective of creating markets free from political control, ministers lose the ability to control the process of reform, but simultaneously other actors will be using the opportunity to clearly attach responsibility to the ministers. Fourth, the conflicting objectives of market reforms, which give some actors the responsibility for equity and others the responsibility for efficiency, create distinctions that are artificial and difficult to clearly and consistently follow. Implementation thus leaves reformers, and their policies, vulnerable. Voters may react negatively to the concept of markets in health care, priming their reaction as reforms are put in place. “On the ground” problems in health care reform and negative news stories tend to be generalized into causal stories that taint reforms and reformers. The essay is structured around four sections. The first section introduces economic theories that drove market-oriented reforms in European national health service systems. This is followed by an analysis of reform implementation in New Zealand and a section explaining the reform outcome. In the fourth and final section, the two hypotheses are compared with the European experience. Comparative evidence shows similar dynamics operating in all national health systems, while finer-grained analysis of the New Zealand case helps us to understand the roots of policy reforms that diffused in all NHS countries. A number of different data sources provide a rich source of evidence for the New Zealand case study. Data are drawn from many public opinion surveys, election studies, and newspapers. Archival documents reveal how political executives anticipate and respond to public reaction to market reforms. This allows us to understand reforms from the perspectives of multiple actors, mediated by the press. Ideas, Problems, and Solutions: The Origins of Market Reforms in European NHS Systems

Reforms in Italy, Spain, Sweden, and the United Kingdom sought to address multiple problems perceived in health care that were defined and proposed through the lens of economic theory. The framing of problems in the health sector primarily as efficiency problems was influenced by changing ideas in economic theory and a belief in the superiority of mar-



1068


ket mechanisms over government. Two major ideas underlie many of the health care reforms: a belief that agency theory (discussed below) could help resolve market failures in health care and that managed competition, rather than hierarchies or planning, was the best mechanism for efficient resource allocation. These ideas were translated into five policies: (1) increased cost sharing for patients through user fees, (2) the separation of purchaser-provider functions, (3) management reforms of hospitals, sometimes paired with profit or return on capital expectations, (4) provider competition, and (5) vouchers for purchasing health insurance. The theories are defined, and each of the five policy responses is briefly described. The implementation of these five policies in Italy, Spain, Sweden, and the United Kingdom is then explained. Ideas: Agency Theory and Managed Competition

Agency theory or principal-agent theory (see Alchain and Demsetz 1972; Jensen and Meckling 1976) suggests that actors in an economic transaction have fundamentally different interests, and as a result, agents do not always act in the best interests of principals. Agency theory is well suited for understanding health care, where (traditionally) providers make decisions independent of third-party payment. Reforms based on agency theory will clearly specify the relationships between agents and principals. Policy reforms will seek to align incentives so that the agent will pursue the objective of the principal more closely. In the United States, the development of managed care was one response to the agency problems inherent in third-party reimbursement. In Europe, consumers and hospitals were the targets for three agency-inspired reforms: consumer co-payments, purchaser-provider splits, and the organizational reform of hospitals. At the level of the health care consumer, agency theory suggests that users who can access free health care have fewer incentives to utilize services appropriately. The policy response is to propose modest user fees for primary care, inpatient care, and pharmaceuticals. However, in Europe in the 1990s, the scope of introduction was typically less than policy makers initially proposed, with all countries adopting less comprehensive user fees than first proposed. A compelling application of agency theory is the separation of purchasers and providers and the management reforms of hospitals. NHS or integrated systems have been faulted for having few incentives to minimize



Laugesen

■


unit costs, since budgets were often reduced if money was underspent (OECD 1992), and productivity and customer responsiveness are also seen as problems because providers held geographic monopolies. For example, county councils in Sweden and local health units in Italy both determine the services to be provided and the hospitals at which they are provided. The purpose of the purchaser-provider split is typically to allow service contracts to be directed to the most efficient providers, or mix of services, rather than reflecting past investments in specific institutions. Competition between providers is not necessarily present, though: a softer version of the purchaser-provider split simply gives one organization formalized power and contracts with providers. In all countries except the United Kingdom, this involved moving from global budgets to a diagnosis-related group payment method or a combination of both. In the four European countries studied, hospitals were given more independence and encouraged to adopt private sector methods of management. Sometimes this required hospitals to meet targets that were drawn from the private sector. The United Kingdom perhaps went farthest, by requiring trusts to make a return on capital. Management reforms also tried to give hospitals greater operating autonomy while developing performance standards. The second idea that has been influential in market reforms is theories of competition between insurers replacing or offering an alternative to the NHS systems. Competition between insurers can increase consumer choice and efficiency. However, Alain Enthoven considered that unmanaged insurance market competition would cause numerous market failures and therefore he also proposed regulation to protect consumers. Standard benefits packages, including the possibility of identical co-payments; annual enrollment; continuity of coverage; oversight of insurers; and quality assurance are some of the safeguards proposed (Enthoven 1988). The theory of managed competition was reflected in health care plans proposed in Italy, Spain, and the United Kingdom, but these plans were quickly shelved (more discussion on this follows in the final section). The major difference between most market reforms and the Enthoven theory of managed competition was that reformers in NHS systems conceived of individuals as the purchasers. Enthoven advocates institutional entities, either the federal or the state governments, as the purchasers. The principles of competition were extended to providers. Some analysts suggest that the policy of the internal market and purchaser-provider split was primarily encouraged to stimulate provider competition (JérômeForget, White, and Wiener 1994). In the United Kingdom, competition



1070


was proposed as a new weapon allowing government to “replace years of fighting well-entrenched professional and institutional powers using administrative orders with market forces that would make providers compete to make health care more affordable” (Light 2001a: 1157). Next, the application of agency theory, or competition, is discussed in Italy, Spain, Sweden, and the United Kingdom. Problems and Solutions: Reforms in Italy, Spain, Sweden, and the United Kingdom

In Italy, market health care reforms were implemented along with major changes to the role of the center and the localities. After 1992, regions exchanged more responsibility in health care for a hard budget cap on the regions’ expenditure, with hospitals restructured as public enterprises (France and Taroni 2005). Prior to reform, local health authorities were governed by elected representatives, and these organizations approximated the roles of the district health authorities of the United Kingdom (Donatini et al. 2001). Local health units previously were subject to considerable manipulation by political parties, since they were used for patronage purposes (France and Taroni 2005). Local health authorities customarily overspent their budgets (ibid.), but were funded nationally, so they faced no specific incentives to be prudent. After 1992, local health units and public hospital trusts were created and managed by an appointed general manager and were given more autonomy along with performance expectations (Donatini et al. 2001). Today, local health unit services are allocated budgets by the regions, under a global budget with a weighted capitation mechanism, and hospital providers are paid fees for services based on diagnosis-related groups for inpatient care. Italian health care reforms differ from the United Kingdom because of a less rigid separation between purchasers and providers from the start of the reforms in 1992; contracts were also not the dominant mechanism to negotiate price, volume, and cost, with a prospective payment system chosen instead. Providers compete for patients not just in their regions but also in other regions, since other regions are required to pay for care elsewhere. Local health units can contract with private providers, as long as the provider is accredited. A proposal to allow citizens to opt out of the NHS that was passed in 1992 was reversed in 1993. Several unsuccessful attempts were also made to introduce co-payments for hospital services during the 1990s (ibid.). Spain’s national health service emerged through the 1978 constitution, although its implementation was largely shaped by the 1986 General Health Act. Regional governments are primarily responsible for the opera-



Laugesen

■


tion of the health care system. By the start of the 1990s, policy makers were concerned about rapid growth in expenditure, increasing waiting lists, and mounting user dissatisfaction (Rico and Sabes 2000). In 1991, a commission (in the April report) suggested cost-containment and managerial reform proposals, but these were interpreted by the public as proposals for privatization and shelved. Regional governments and later legislation developed some of those policy proposals, including changes to the organization and management of hospitals, experimentation with contracts and prospective payment systems, the launching of several measures to contain pharmaceutical costs, and small steps toward increasing the role of private management and ownership in the public sector (ibid.). Co-payments have not been introduced in Spain, except for pharmaceuticals, since past discussions on co-payments have generated widespread public opposition (Gaminde 1999). Hospitals have undergone a range of varied conversions, including public enterprises, trusts, consortiums, and foundations designed to introduce “flexibility and to ease up the rigidities involved under the present organizational framework” (Rodriguez, Scheffler, and Agnew 2000: 120). Some private sector services are contracted out with the public system, but there is no competition among providers (public or private) (Rico and Sabes 2000). Proposals to shift the financing system to one of managed competition have been very unsuccessful in Spain. Sweden’s twenty-one county councils finance and manage the majority of health care services in Sweden. Inefficiency, waiting lists, and patient responsiveness, as well as fiscal pressures, were issues that prompted reforms (Anell 1996). The 1985 Dagmar Reform introduced the purchaserprovider split, which was adopted in some counties. Planned markets were introduced in 1988 and in most counties focused on patient choice of provider and managerial independence of hospitals, but they were not uniformly adopted (Saltman and Bergman 2005). The separation of providers and purchasers was weak compared to the separation in the United Kingdom, and approaches to competition varied from invitations for providers to tender, whereas others worked to coordinate providers and improve health outcomes (Anell 1996). Accredited private providers can receive public reimbursement for outpatient care, but the role of private providers remains relatively small in Sweden. In general, county facilities operate as monopolies within each county, although more counties allowed patients to choose their own hospital in the 1990s (ibid.). In the United Kingdom, an internal market (1991–1997) was created in response to a fiscal crisis in the late 1980s. Policy makers had a model for the health system that emphasized competitive markets, based on their experience with privatizations in other sectors (Robinson and Dixon



1072


1999). However, ministers were reluctant to change the financing of the system; attention shifted to the way services were organized, managed, and delivered (ibid.). The rationale for the purchaser-provider split was partly related to the lack of incentives for increased productivity at the hospital level: if providers treated more patients, their income did not increase (Ham 1996). In the United Kingdom, a metaphor used repeatedly was that money should “follow the patient” or that providers should receive funding based on the treatment they actually provide to reward providers who delivered well-managed care and who were responsive. District health authorities could establish secondary care contracts with the hospitals of their choice and not simply those they managed directly (ibid.). In the primary sector, fund-holding was established for primary care purchasing. Fund-holders were primary care practices responsible for purchasing enrolled patients’ health care, including hospital care, a policy unique to the United Kingdom. Hospitals were converted to trusts and expected to make a return on their capital value. As mentioned, ministers were reluctant to introduce a system of vouchers for insurance, mainly because of the opposition this option generated. This summary of market reforms shows some commonalities. Among national health service models, there was a degree of uniformity in the timing (between 1988 and 1992) of reform, the negative fiscal conditions that seemed to encourage reforms designed to transfer risk to users, and the reasons for reform. Policy reforms were proposed to solve similar problems of inefficiency, consumer responsiveness, and productivity in all four countries. The reforms also show considerable uniformity, with user fees, purchaser-provider separations, and hospital management reforms. Public opposition in all cases stalled more proposed financing reforms and delayed implementation of other policies in Italy and Spain. Provider competition and private sector provision is variable and plays a minor role in most of these countries. In Sweden and Spain, the local and federal political structures hampered efforts at the central level to impose uniformity, but in both cases, some regions or counties implemented more market-oriented purchaser separations and hospital management reforms. Next, the implementation of these policies in New Zealand is explored. Market Reform of Health Care in New Zealand

The introduction of market policies in health care in 1991 followed, and was influenced by, seven years of economic reforms introduced to resolve



Laugesen

■


New Zealand’s indebtedness and declining economic growth. Trade was substantially liberalized, employment law was reformed, and agricultural subsidies were abolished in the 1980s (for an account of economic reforms, including the privatizations discussed below, see Evans et al. 1996). Agency theory was the basis for many of New Zealand’s economic and public sector reforms in the 1980s (Boston 1991b). The most important economic reform to influence health care later was the government transformation of government-owned banking, energy, telecommunications, and transportation businesses. Many of these businesses were privatized by the mid-1990s. Between their prereform status as government departments and their later status as fully privatized companies, these organizations were corporatized or were restructured as quasi-commercial state-owned enterprises. Under the State Owned Enterprises Act 1986, these businesses were required to be “as profitable and efficient as comparable businesses that are not owned by the Crown” (Section 4–1a) as well as “a good employer” (Section 4–1b) and “an organisation that exhibits a sense of social responsibility by having regard to the interests of the community in which it operates” (Section 4–1c). This legislation provided the template for hospital reform in 1992, with some adaptations. Health care reforms in 1986–1988 that used a corporatization model (Hospital and Related Services Taskforce 1988) failed, although the government did succeed in regionalizing the health care system. By 1991, fourteen area health boards provided hospital and local public health services. Elected and appointed representatives governed these boards. The minister of health allocated global budgets to area health boards based on population characteristics within each board’s boundaries. Physicians in private practice delivered most of the primary care services and were paid through physician, laboratory, and pharmaceutical benefits administered at the national level. Generally, physicians worked on a fee-for-service basis. Hospitals provided free emergency care and free nonemergency care for patients with a referral from a primary-care physician. Various problems identified in the 1980s were a stimulus for reform. Primary care was not as affordable for patients of low-socioeconomic status because the fee-for-service fee subsidy from the government averaged around 20–30 percent for physician visits (Fougere, Scott, and Marwick 1986), meaning the average co-insurance was 70–80 percent for a consultation. There were long waiting lists for surgical procedures in the hospital sector, but meanwhile, 25 percent of procedures were performed in private hospitals (ibid.). Poor financial management and inefficient use of



1074


resources characterized the hospital sector (Hospital and Related Services Taskforce 1988). As the discussion below shows, reformers in the early 1990s drew on these themes in their critique of the health care system. Labour lost the 1990 election after losing support from its traditional base of voters. Labour’s constituency of low-income and unionized voters and social liberals (see Nagel 1998) faced contradictory preferences on economic reform, with the displaced workers and government employees tending to support the economic reforms less readily. The National Party was elected to the government in 1990, and soon after, the cabinet announced reductions to social welfare benefits and promised to reform the health care system. Health reform proposals were developed quickly in order to ensure interest groups could not direct reform. Proposals were developed between December 1990 and June 1991 by a taskforce of bureaucrats. Reformers did not examine a wide range of policy options outside of market reform, with almost all options but one status quo option involving competition in provision or financing (Ministry of Health Files 1991a). Around seven different possible structures were proposed in March 1991, and the options were developed into a single proposal by May 1991. The May proposal was the basis for the reform plan, which contained all of the common elements of market reform identified earlier: (1) increased cost sharing for patients, (2) a separation of purchaser-provider functions, (3) managerial independence of hospitals, (4) provider competition, and (5) vouchers for purchasing health insurance. After announcing the reform in mid-1991, the minister of health replaced elected hospital boards with political appointees. Two reform laws were passed almost immediately in 1991. These were separate pieces of legislation because both were amendments to different existing legislation: the area health boards were under the health boards legislation while co-payments required amendments to social security legislation. More comprehensive reform legislation was introduced in August 1992 and passed in 1993. The minister of health was advised to make policy changes swiftly and replace board members at the time of the announcement of reform, because of the risk that board members would obstruct reform (Ministry of Health Files 1991c). Elected representatives were thought to be unable to implement change, because of “the politicization of the decision making process” (Upton 1991: 9) and responsiveness to interest groups and geographic constituencies rather than broader communities (9). A purchaser-provider split was proposed to increase efficiency: “The conflicting purchasing and



Laugesen

■


provision roles of area health boards hamper the quest for value for money in the health system, and make it impossible to reward the most effective health care providers. The two roles should be split” (23). The key problem facing area health boards is that they are expected to be both purchaser and provider of services. These roles conflict, making it difficult for boards to carry out either role effectively. This leads to other problems that are more visible to the public, such as long waiting times for hospital care and lack of responsiveness to individuals’ needs (ibid.). Four regional health authorities were created to purchase services from public and private hospitals. These regional purchasers were expected to encourage hospital competition (between public and private hospitals and between public hospitals) because they were no longer compelled to buy services from public hospitals. Hospitals and area health boards were renamed crown health enterprises (CHEs) in 1993 and were established to compete with each other for contracts from the regional purchasers. CHEs were expected to be as successful as nongovernment businesses, and they were also expected to be socially responsible and to consider the interests of the community where they were based. A new ministerial portfolio, the minister for crown health enterprises, was established. To behave more like private businesses, hospitals would have increased autonomy and hierarchical management models would be replaced by smaller, independent organizations. It was hoped that this would ensure greater flexibility in the kinds of services available (Press [Christchurch] 1992) along with greater consumer responsiveness (Bay of Plenty Times 1992; Manukau Courier 1992). The government said it would introduce vouchers for health insurance once regional health authorities were established (Upton 1991). Consumers could opt out of the government insurance plan once regional health authorities were operational, allowing citizens to use a voucher. Insurance companies could compete for enrollees. However, a voucher system for health insurance was recognized privately early on as requiring a long and costly implementation process that would be unpopular with voters (Ministry of Health Files 1991e). A standard health benefits package was also proposed to regulate insurers’ provision of services and to encourage ranking of different services. Reformers thought the rationale for introducing user charges and integrating the overall funding for health care was to reduce “fiscal risk” (Ministry of Health Files 1992c: 3) and to provide a way to “involve the patient in decisions which influence the utilization of health care more effectively by giving him or her a financial stake in that decision” (4).



1076


Once patients were acting as their own agents and buying social insurance, consumers would “signal how much health expenditure [an individual] thinks is appropriate,” which would result in “better incentives for self care, cost-effective care and saving for the health care costs of old age” (Ministry of Health Files 1991b: n.p.). In the view of advisers in the Health Reform Taskforce, risk-adjusted social insurance premiums would “encourage people to live healthier lives” (ibid.). Authors suggested that risk-adjusted social insurance premiums would result in savings because “individuals would have a greater incentive to live healthy lives and avoid accidents, economize on their use of health services and save for the health costs of old age” (Ministry of Health Files 1991d: n.p.). New Zealand reformers thought that individuals would be more careful with their health and safety, and that they would use services more appropriately, if they bore more of the cost of utilizing services directly. The impact of the market reforms in New Zealand was an increase in private expenditure and a slowing in public expenditure growth. Perperson private expenditure increased from NZ$289 in 1990 to NZ$433 in 1996. This was an increase of NZ$149, or about USD$100, in constant dollars. In contrast, per-person public expenditure on health care grew only marginally, by only NZ$67 between 1990 and 1996, in constant dollars. Private health expenditure as a percentage of total health expenditure rose from 17.6 percent of total health expenditure in 1990 to 23.3 percent of total health expenditure in 1996, declining marginally to 22 percent in 2000 (OECD 2001). Many reforms were partially implemented or reversed between 1993 and 1999. For example, against advice from Treasury and other officials, between 1992 and 1993 ministers retained an interim co-payment and targeting regime. The minister of health said he did not wish the new scheme of targeting and co-payments to generate “losers” in 1993 (the election year). Officials suggested that the only way to avoid generating losers in 1993 would be to retain the interim targeting regime until early 1994 (Ministry of Health Files 1992b). The cabinet decided at the end of 1992 that problems of access to pharmaceutical products for low-income individuals justified a number of immediate changes, including the option of an overall dollar stop loss for pharmaceutical products (Ministry of Health Files 1992i). Bureaucrats saw these changes as politically influenced (Ministry of Health Files 1992d). Maternity care and accident treatment were exempted from charges, and imaging and laboratory benefits were reinstated (Ministry of Health Files 1992g, 1992k). Government officials advised against changing the system of co-payments



Laugesen

■


and means tests for a number of reasons. Financial estimates had been made on the basis of continued operation of the scheme (National Archives 1991). Treasury was against making further changes to the means-testing policies as changes would erode credibility and reduce the ability of the government to increase efficiency and expenditure (Ministry of Health Files 1992f), and Treasury also did not want to see the scheme expanded lest it create more demands for change. Treasury advised the minister not to modify the targeting regime because modifications would serve to “fan public debate on its necessity” (Ministry of Health Files 1992j). Officials suggested devolution of user charges could reduce political interference (Ministry of Health Files 1992d). CHEs, the former public hospitals, were supposed to be successful and socially responsible, but in practice, after they were created, they focused much less on social objectives and more on profitability, probably because business and financial performance were more important to government than social objectives. Around 96 percent of the CHE directors thought business planning was a very high or high priority for government, closely followed by financial performance (86.3 percent). Only 33.7 percent of respondents considered social responsibility to be a very high or high priority for government (Barnett, Perkins, and Powell 2001: 148). The idea of developing a standardized package of core services was dropped, partly because of the negative attention surrounding defining the package and partly because it was less important in the absence of competing health care plans. It was replaced with a greater emphasis on priority setting, development of guidelines, and access based on clinical criteria. Competition between regional health authorities and vouchers for health insurance was not implemented, mostly because of the public reaction to these proposals (Laugesen 2001). The election in 1996 triggered further modification to the National Party’s health policies. Institutional factors played a role in the demise of these policies, since the introduction of proportional representation in 1996 increased the competition between parties and made smaller parties more powerful. National signed a coalition agreement with the New Zealand First Party, a party that had criticized the health reforms. The agreement with New Zealand First finally doomed National’s market reform policy. Competition between hospitals and the idea that hospitals would be forced to close if they were unprofitable never gained traction and were abandoned in 1996. Public hospitals were no longer expected to be profitable (Steering Group 1997: Appendix II). After this, public hospitals were recast as mostly regional monopolies. The crown health enterprise label



1078


was eliminated altogether, and CHEs were renamed hospital and health services. Under the terms of the postelection 1996 coalition agreement (ibid.), regional purchasers were abolished and their functions transferred to one national purchasing agency. Throughout the period between 1996 and 1999 the minister of health emphasized to the public that the competition model had been rejected (English 1998) and that it could expect more stability in the health sector overall (Ministry of Health 1999) rather than constant change. After 1996, when some of the harsher aspects of the internal market were reversed, public financing of health care increased while private financing decreased. In the three years between 1996 and 1999, total per-person public expenditure grew by NZ$222. Following the coalition agreement, between 1996 and 1999 real private expenditure grew much more slowly, by only NZ$45 (Organisation for Economic Co-operation and Development 2001). Then, in 1999, a center-left Labour-Alliance Coalition of Maori (the indigenous people of New Zealand) and green parties was elected. The minister of health said she wanted to establish a health service based on “co-operation and collaboration” to replace “the current commercial and competitive model” (Evening Post 2000). The New Zealand Public Health and Disability Act 2000 appeared to rebuild the 1991 structure. This seemed an effective way for the coalition to symbolize to voters that the market model was buried. Some described it as “back to the future” (Devlin, Maynard, and Mays 2001) reform. Twenty-one district health boards were established in October 2001. These boards have a mix of elected and appointed members, similar to the area health boards in existence in 1991. Boards have seven elected members and four appointed by the minister of health, based on both skills and representation, with significant attention to Maori representation. Boards act as both purchasers and providers of hospital services (New Zealand Public Health and Disability Act 2000, Section 1). The Partial Survival of Market Reform and Explanations

In all the countries discussed, public reactions muted extensive reform of the private sector. For example, in Galicia, a province of Spain, a diverse group of citizens, unionists, parties, professionals, and others developed a highly effective campaign against a privatization effort that then spread to the rest of the country (Bayle and Cal 2001). Galicia’s health reforms



Laugesen

■


were strikingly similar to New Zealand’s reforms, and the public reaction was similar, despite the very different cultural, political, and historical context of reform in both countries. This suggests that the public finds market reforms of health care to be problematic. The hypotheses explored below are (1) that health care is different from other areas of public policy and that voters will tend to resist policies of market reform in health care and (2) that while user fees are a form of retrenchment, market reforms in national health services are distinct from retrenchment, because of their complexity and potential for counteroffense by interest groups. Market Reforms in Health Compared to Other Market Reforms

Policy makers in New Zealand learned lessons from other areas of the economy and public sector reform and applied them to health care. As mentioned earlier, the State Owned Enterprises Act 1986 was applied to government postal, transportation, energy, and telecommunications enterprises, and this provided the major model for restructuring of hospitals into crown health enterprises. The first version of the Health and Disability Services Act copied clauses from the State Owned Enterprises Act: the act required state-owned enterprises to be as profitable and efficient as nongovernment businesses, to be a good employer, and to have a sense of social responsibility. Other aspects of the reforms such as the purchaserprovider split reflected reforms in the United Kingdom, but also reflected a broader set of principles recommended by the Treasury Department in New Zealand in the late 1980s. Treasury recommended the separation of commercial and noncommercial activities, separation of the responsibility for the provision of policy advice, regulatory and funding activities from operational activities, and devolution and contracting out of policy implementation (Boston 1991a). Ministers became purchasers of policy advice (potentially from a range of sources), and programmatic functions (such as delivering unemployment benefits) were separated from the policy functions. Much of the latter managerial changes were invisible to the median voter. But when these ideas were applied to health care, the reception by the public was altogether different. This may reflect underlying differences in attitudes to health. In the United States, voters consider health care different from other social welfare policies (Schlesinger and Lee 1993). Health care in the United States is associated with support for equal opportunity, whereas more overtly redistributive policies are asso-



1080


Table 1

Private and Government Operation of Services, 1997 Electricity

Mainly run by private organizations Mainly run by government Don’t know

48% 46% 6% 100%

Banksa 74% 17% 8% 99%

Hospitals 18% 78% 4% 100%

Source: New Zealand Role of Government Survey 1997. a Due to rounding, percentages do not total 100%. Number of responses: electricity 1,182; banks 1,174; hospitals 1,183.

ciated with the characteristics of the recipients of the services. People are less tolerant when it comes to redistributive public programs for the poor and more tolerant of health care programs (ibid.). In countries outside of the United States that have universal insurance, health care might be thought of as less different than social welfare policies but different from other areas of the economy. To understand reactions to reforms that emphasize greater use of markets, the most instructive comparison is similar reform policies in other policy areas. Support for heath policies is compared with support for economic policies, rather than social policy, as Schlesinger and Lee did. If health care is different then voters are less likely to support market-oriented health care reform to the same extent as they support market reforms in other areas. To test the hypothesis of whether health care is different from other market reforms, opinions on health care were compared with opinions on ownership of former state-owned enterprises, and public opinion trends in support for government expenditure were also compared over time. Data on attitudes to private or public ownership in health and other sectors were compared. A question in the New Zealand 1997 Role of Government Survey asked respondents whether they thought hospitals, banks, and electricity should be run mainly by the public or private sectors. Table 1 shows that support for private operation of electricity was 48 percent and support for private operation of banking was 74 percent. Support for private operation of hospitals was only 18 percent. This lends support for the idea that attitudes to ownership of hospitals are considerably more supportive of government ownership compared to the banking and electricity sectors. Data from three rounds of the New Zealand Election Studies were compiled for the years 1993, 1996, and 1999. The New Zealand Election Study (NZES) is a national postelection postal survey. Table 2 reports data on support for ownership of specific companies that had been fully



Laugesen

■


Table 2 Opinions on Ownership of Former Government Businesses, 1990–1999

Preferences for ownership More public 100% public ownership Partial public ownership More private Private ownership with regulation Private ownership, no regulation Don’t know Total Respondents

Electricorp 1993 1996 1999

Bank of New Zealand

Telecom

1993 1996 1999 1993 1996 1999

54% 39%

40% 39%

10%

27% 34%

27% 33%

17% 21%

19% 22%

27%

22% 19%

20% 18%

17% 26%

28% 20%

22%

32% 16%

20% 20%

6% 8% 9% 12% 25% 14% 23% 28% 21% 6% 5% 5% 6% 16% 5% 4% 6% 8% 100% 99% 101% 99% 100% 100% 96% 101% 100% 2,034 4,017 5,110 2,029 4,262 5,133 2,021 3,983 5,106

Source: New Zealand Election Studies, 1993, 1996, 1999 Note: Percentages do not total 100% because of rounding.

or wholly privatized (Electricorp, Telecom New Zealand, Bank of New Zealand). Support for full government ownership of Electricorp declined from 54 to 40 percent in 1999. However, support for partial ownership of Electricorp remained relatively constant, from 17 percent in 1993 to 21 percent in 1996 and 19 percent in 1999. Support for regulation instead of ownership increased from 17 percent in 1993 to 26 percent in 1996 and 28 percent in 1999. The proportion of voters who favored unregulated privatization (neither regulation nor ownership) increased from 6 to 9 percent. Support for full ownership of Telecom New Zealand fell dramatically from 39 to 10 percent in 1996, but then increased to 27 percent by 1999. Support for neither ownership nor regulation increased from 15 percent in 1993 to 25 percent by 1996. The Bank of New Zealand was privatized in 1992 (New Zealand Treasury 2005). When asked about the Bank of New Zealand, respondents’ support for public ownership was steady at 33 percent in 1999. These data suggest that New Zealanders were not wholly supportive of complete privatization, but they did suggest a slight softening in attitudes to privatization. Next, data on the government’s role in health expenditure were assembled, with the purpose of illustrating the public’s perception of how



1082


Table 3

Free Health for All

Definitely should Should Should not be Definitely should not be Don’t know Total

1993

1996

1999

4% 35% 40% 18% 2% 99%

33% 36% 24% 4% 3% 100%

30% 36% 25% 5% 4% 100%

Source: New Zealand Election Studies 1993, 1996, and 1999 Notes: Question: On the whole do you think it should be or should not be the government’s responsibility to provide free health care for all? Number of respondents: 1993, 2,037; 1996, 3,966; 1999, 5,137. The 1993 total does not equal 100% due to rounding.

Table 4

Preferences on Government Spending on Health, 1990–1999

Spending on health

1990

1993

1996

1999

Much more More Same as now Less Much less Don’t know

30% 41% 25% 1% 1% 3% 101%

28% 50% 18% 2% 0% 1% 100%

38% 47% 13% 1% 0% 1% 100%

31% 52% 15% 1% 0% 1% 100%

Source: New Zealand Election Studies 1990, 1993, 1996, and 1999 Notes: Question: Listed below are various areas of government spending. Please show whether you would like to see more or less government spending in each area. Remember that if you say “more” or “much more,” it might require a tax increase to pay for it. Totals for 1990 and 1993 do not equal 100% due to rounding. Number of respondents: 1990, 1,865; 1993, 2,040; 1996, 4,387; 1999, 5,122.

much financial risk the users and government should assume in health care. The answers to two questions in the NZES were compared: whether the government should be required to provide free health care for everyone (table 3) and support for increased public expenditure (table 4). Table 3 shows that the percentage of respondents who thought that “free health care definitely should be provided” increased from 4 to 33 percent between 1993 and 1996, an increase of 29 percentage points. What is striking about these changes in opinion is that they did not coincide with policy changes— reductions in user charges occurred largely prior to the 1993 election. In 1999, a similar proportion (30 percent) of respondents thought that the government definitely should provide free health care to all. There was a decline in support for individual responsibility for



Laugesen

■


health care, as measured by the change in the proportion of respondents who thought the government should not provide free health care. The percentage of respondents considering that “the government shouldn’t provide free health care for all” decreased from 40 percent in 1993 to 24 percent in 1996 to 25 percent in 1999. The number of staunch opponents of government expenditure increases (those who said “the government definitely shouldn’t provide free health care for all”) declined from 18 percent of the population in 1993 to 5 percent in 1999. Respondents in the NZES were asked whether they thought the government should spend much more, more, the same, less, or much less on health care (shown in table 4). Respondents were told that greater expenditure could cause tax increases. This question’s wording asks respondents to consider the trade-off of increased government expenditures. Support for much more expenditure hovered around 30 percent throughout the 1990s, peaking in 1996 at 38 percent, but support for more government health expenditure increased from 41 percent in 1990 to 52 percent in 1999. Support for keeping the same level of expenditure declined from 25 percent to 15 percent. The survey data show that responses to market reform in the health sector were different from responses to market reform in other sectors and that voters wanted a greater commitment to government funding for health care and clearly supported public operation of hospitals. Earlier it was hypothesized that health care is considered similar to other social policies (whereas it is different in the United States, according to Schlesinger and Lee 1993). One clue is given from the public submissions sent to the Parliamentary Committee on Social Services when the health reform bill was before Parliament. Groups and individuals who made submissions on the crown health enterprises disagreed with the legislation’s objective for crown health enterprises’ business success, suggesting the provision of health services should be the primary objectives of these new organizations (Department of Prime Minister and Cabinet 1992). Submissions expressed concerns regarding the effects of market restructuring and splitting former boards into competing hospitals. This process was thought to put existing cooperative relationships between hospitals within a district at risk. Others considered that profit is “not a relevant indicator of efficiency in the delivery of health services” (ibid.: 41). Many of the submissions considered that meeting both health and business goals is incompatible and that health is not a commodity (ibid.). These factors indicate some of the differences between the reform model of health care as a service like others and the public understanding of



1084


health care as having unique characteristics. Likewise, as the next section shows, in table 6, the kinds of expectations that the public associated with reforms included impersonal or businesslike perceptions of the health care system, along with a palpable pessimism about the actual quality of postmarket health care as “going downhill.” The public also predicted that the health system would be “more expensive,” that spending priorities would be directed to administration rather than patients (“more money on administration, less on patients”), and that “smaller districts will have to look after themselves” (Ministry of Health Files 1993b). Blameworthy Features of Market Reform

When governments make cuts to the welfare state, they must engage in various strategies to obfuscate policy reforms, since the reforms are usually very unpopular (Pierson 1994). Retrenchment is a distinctive and difficult political enterprise and it is a blame-avoidance exercise (rather than a credit-claiming exercise), because of resentment from losers. Governments will use the same political strategies used in retrenchment to avoid blame for health care reform (Light 2001a). Blame is exacerbated by the fact that in a national health service, the government gets blamed for something a manager or doctor does in some part of the system (ibid.). Blame does drive the trajectory of market health care reform. At the same time there are similarities in political strategies between classic retrenchment processes and market reforms, there are also some important differences. Whereas retrenchment cuts benefits, market reforms alter the very fabric of the health service and create new institutions. First, the scale is larger, in terms of resources and people, and the task of encouraging formerly bureaucratic organizations to embrace competition is challenging. Second, professionals occupy a different position and are a sympathetic group that can garner the support of public opinion. Third, in pursuing the objective of creating markets free from political control, ministers lose the ability to control the process of reform, but simultaneously other actors will be using the opportunity to clearly attach responsibility to the ministers. The Opposition plays a significant role in all political systems in highlighting reform deficiencies. Fourth, the conflicting objectives of market reforms, which give some organizations the responsibility for equity (as was the case with the regional purchasers in New Zealand) and others the responsibility for efficiency (the crown health enterprises or former public hospitals), can be difficult to clearly and consistently follow.



Laugesen

■


Since market metaphors may not mix with health care, it is also possible that these views prime voters for negative reception of reform implementation. Regardless, the disruption caused by “making a market” and deregulating the health care sector leaves reformers, and their policies, extremely vulnerable, beyond the shifting of costs to specific actors. Voters see some situations as the responsibility of government or other actors (Stone 1989). The discussion below explores why voters attribute blame to the government. Generally speaking, blame often is “an immediate, visceral, emotion-laden reaction to a set of events that does not depend on careful or extended reflection” (Rosenthal and Schlesinger 2002: 49). Generally, a large proportion of the public is unaware of the details of health care reform. However, cognitive psychology suggests that people use limited informational capabilities to create shortcuts (Iyengar 1990) and people tend to use schema to organize information cognitively (Khong 1992). People demonstrate a tendency toward risk aversion in their choices (Kahneman and Tversky 1984). This is borne out in the data on awareness of reform, where 38 percent of respondents had not heard anything about the health reforms in New Zealand, but those that could recall information understood overnight hospital charges, prescription charges, payment issues, and closure of hospitals (Ministry of Health Files 1993b). Voters understood commercialization through media reports of high salaries and benefits (designed to attract new people to jobs in hospital management). Another easily understood schema around national health service reforms is that health care will be Americanized and that people will be no longer able to seek free care in the hospital system but must adopt new health insurance. The Americanization comparison came up frequently in the media (and the media database described below) because it represented an easily communicated vision of postreform health care by opponents of market reform. The need for insurance, in a system in which citizens get care and see hospitals as the symbols of the national health service (especially perhaps in the United Kingdom), the name changes, and the commercial focus of hospitals represent a substantial shift in the health care infrastructure that people are familiar with. Although the public was concerned about access and quality (Steering Group 1997), it was the more commercial aspects of reform that seemed to create the most anxiety. The public was concerned about the profit objectives of CHEs, whether health care was going to be privatized, and if they would have to take individual responsibility for financing health care (Ministry of Health 1996). This is the background against which the reform experience is projected. Voters see some situations as not amenable to human intervention, but



1086


others are seen or come to be seen as the responsibility of government or other actors (Stone 1989). Voters generally have a negativity bias (Weaver 1986). Under conditions of retrenchment we might expect causal stories about who is to blame to tend to stick to the government. Causal stories about the transition to market are likely to be negative. Market signals and incentives may take a long time to develop, in an integrated health care system. Different actors must pursue different objectives, objectives that sometimes conflict or are not as clear-cut as reformers first envisaged. Negative causal stories link problems such as patient safety failures or waiting lists to the implementation of reform, and voters notice when outof-pocket costs go up or when public hospitals cut unprofitable services. Market reform, with its emphasis on multiple organizations pursuing different objectives, makes coordination between government actors very difficult. A microcosm of this process was evident in the first two years after reform was announced, when there was poor communication and “relationship difficulties” between CHEs and Regional Health Authorities (RHAs) (Ministry of Health Files 1993c:4). Even those at the center of the policy process complained about the impact of the commercial model on their effectiveness. The chief executive of the Ministry of Health complained that it was difficult to coordinate with so many central agencies responsible for the health sector, and the ministry could not get access to CHE business plans without signing confidentiality agreements (Ministry of Health Files 1993a). These features of market reform, their effect on voters, and the reactions of the government to these consequences shape the public’s understanding of the progress of market reforms. The “on the ground” problems in health care reform tend to get interpreted as causal stories about the nature of reform more generally, even if voters do not understand health reform. Political institutions are important for influencing the course of causal stories. Westminster systems tend to centralize blame (Weaver 1986). Just as important is the role of opposition parties. Prior to 1996, New Zealand was basically a two-party system, and these norms survived in the multiparty era. Causal stories are taken up or invented by interest groups and opposition parties in Westminster political systems. The vigorous criticism by the Opposition in Parliament has the effect of continual scrutiny of reform that can chip away at reform implementation. If voters do not hold the minister of health accountable, opposition parties will try to expose flaws and redirect blame back to central government, and health reforms seem to provide endless opportunities for this kind of blame game. Market reforms of health care have some unique characteristics, in



Laugesen

Table 5

■


Approval of Reforms, 1991 and 1993 August 1991

Approval of budget’s effect on health Disapproval of budget’s effect on health Don’t know

October 1991

24% 67% 8%

Do you feel that the changes the government is making will improve the public health system?

10% 80% 10%

March 1993

Yes No Don’t know

11% 75% 14%

Source: Heylen Research Center. Note: August 1991, N= 945; October 1991, N=944; March 1993, N=935

terms of the ability of interest groups to elicit sympathy from voters. In New Zealand, nurses and hospital specialist physicians organized against reforms. It turned out that the media was a receptive arena for these groups. Typically, the people brought in by the government to reform hospitals, for example, were businesspeople from other sectors. Businesspeople lacked credibility on health issues and further served to imply reforms were segues to privatization. Business executives were not as compelling as experienced nurses and physicians. Professional groups shaped the agenda and also communicated effectively: “I believe the public’s reaction to the reforms to date is less dependent upon the lack of information than the influence of the position adopted by the health sector—nurses, GPs, Medical Association, salaried medical staff and so on” (Ministry of Health Files 1993d: 2). As a result, voters had poor impressions of the reforms. Table 5 shows voters’ dislike increased and solidified over time. Immediately following their announcement, 67 percent disapproved of the health reforms, increasing to 80 percent in October 1991. The fact that so many people disapproved of the reforms well before the health care system had significantly changed suggests that the public’s underlying values (health care is different) were incompatible with the reforms’ commercial focus. In March 1993, 75 percent, a substantial majority, disagreed that the changes that the government was making would improve the health care system. A different question, asked closer to the full implementation of the reforms, suggests that the public had not changed its mind or had not been convinced of the merits of the market approach. The Ministry of Health’s own survey of the public (1,000 people in



1088


Table 6 Perceptions of What Is Likely to Happen to the Health System as a Result of the Changes or Health Reforms, May–June 1993 Anticipated outcomes of reform Don’t know Positive Responses Become more efficient Provide better service Funds allocated on population of area Administrators will become more accountable Negative Responses Go downhill Impersonal/businesslike Smaller hospitals will close More expensive More financial cuts Small districts will have to look after themselves Spend money on administration, less on patients Services they offer will get less Bureaucracy will get bigger Staff cuts/less staff CHEs paid highly

% 38.4 6.6 4.4 2.8 2.6 6.4 6.0 5.9 5.8 5.3 5.0 3.5 3.4 3.2 3.1 2.9

Note: The total here exceeds 100% (104.9%). N = 1,000. Source: Ministry of Health (1993b)

June 1993) gives a more detailed qualitative sense of the feelings that voters had about reforms and the associations that voters had with the reforms. On an open-ended question, respondents were asked to name effects of reforms (see table 6). As mentioned already, a substantial number of people did not know what the likely results of health care reforms would be (38 percent). Positive reform results included higher efficiency (6.6 percent), better service (4.4 percent), that population would determine funding of regions (2.8 percent), and that administrators would become more accountable (2.6 percent). Negative perceptions outnumbered the four positive outcomes. The public perceived deterioration (6.4 percent) of the system (the system would “go downhill”), that the system would be more impersonal and businesslike (6 percent), and that there would be closure of smaller hospitals (5.9 percent). Voters also thought health care would be more expensive (5.8 percent), that there would be budget cuts (5.3 percent), that smaller districts would have to “look after themselves” (5 percent), and that more money would be spent on administration and less on patients (3.5 percent).



Laugesen

■


Table 7 Political Party with Health Position Closest to Respondent, 1990–1999 Party closest on health

1990

1993

Labour National Alliance ACTa NZ First None No difference (1990 only) Don’t know Respondents

21% 39% 24% 17% 1818

38% 11% 23% 5% 7% 16% 1923

1996

1999

44% 13% 16% 12% 4% 11% 3736

52% 13% 12% 3% 2% 5% 13% 4852

Source: New Zealand Election Studies. Notes: Question: Which of the parties’ view—Labour, National, the Alliance, ACT, or NZ First—would you say came closest to your own views on each of these issues? 1990 questionnaire gave only two party options. a Data not available previously

As was mentioned earlier, in Westminster systems blame is more readily attributable to the executive. The executive is also closely identified with his or her political party’s policies. To test the effect of blame for the reforms, partisan support for different political parties on health care issues is compared (table 7). The New Zealand Election Study asked voters to identify the party that was closest to their own positions on health care. The party that introduced market-based health care reforms saw a large and permanent drop, while the major opposition party increased its support and retained consistent support through the 1990s. In 1990, 39 percent of voters considered that National was closest to their views on health and 21 percent thought Labour was closest. By 1993, only 11 percent of voters thought National was closest to their own views on health policy. It is likely that most voters (38 percent) transferred their allegiance to Labour. By 1996, more voters (44 percent) thought Labour was closer to their own views. In 1999, 52 percent of voters considered Labour to be closest to their own views on health, while 13 percent considered National closest. Political elites are attuned to public perceptions about the progress and likely success of health policy reforms. Archival evidence suggests that the implementation process in New Zealand was intertwined with blame avoidance strategies that ministers and advisers considered at every step. For example, in 1992 advisers noted that the public had various negative



1090


perceptions of the health reforms, tended to only see the reforms as being associated with increased co-payments, and thought that very little progress had been made: A communications strategy should underpin, and be integrated with, all of the key steps in the reform process. The personal stake in health issues, held by the public, the association of the health reforms with the part charging regime [hospital charges and other increased co-payments], the perception of minimal action on implementation [of] the Budget announcements, and the complexity of the reform structures, all indicate that the reforms, to be implemented effectively, must be carefully worked through and be perceived to be soundly based and well managed. The communication and implementation process cannot be readily separated. (Ministry of Health Files 1992a: n.p.) Officials considered that the public held misperceptions about the establishment of purchasers, such as “the amount the Government is to privatize the health system/duplicate the US system with private health insurance,” that “reforms were being set up so that they could fail,” and that “reforms will result in a two tier health system” (Ministry of Health Files 1992e). In response, elected representatives have a number of strategies available when they are blamed. Politicians generally use specific strategies to cope with blame, including taking full responsibility but suggesting past or present conditions justify action, pleading ignorance, and blaming a committee or hierarchical structure. Elected representatives will also reframe the consequences of the event as having future benefits or present benefits or they are explained through comparison with past problems, other social groups, or a worst-case scenario. Finally, they can reframe the rationale for policy as relating to a principle such as fairness or personal conscience (McGraw 1990). For the most part, the strategy used by reformers in New Zealand was to try to reframe the nature of the reforms through comparison with the past or on principle grounds. Advisers suggested responding to these problems by increasing the awareness of “problems with current system (e.g. We already have rationing and a two-tier system; gross inefficiencies resulting from incentive system in place),” whereas in the future consumers would have more integrated health care and the right to join health care plans. Another strategy was to present the reforms as “incremental and evolutionary” (Ministry of Health Files 1992e). But the practical implementation issues of reform provided no concrete



Laugesen

■


gains for consumers, which added to the credibility problems of the government when it continually emphasized the efficiencies of the market. In the case of the contracting process, rather than quickly providing needed efficiencies in health care, remaking service funding through the use of contracts created problems in the contracting process between regional health authorities and crown health enterprises (the former public hospitals). Hospital enterprises did not have incentives to provide higher volumes or to significantly increase quality of care. Instead, reforms may have encouraged enterprises to build up waiting lists to support their demands for funding increases (Ministry of Health 1996: 30). And while the minister of crown health enterprises was holding these enterprises to profit expectations, the minister of health provided troubled enterprises with deficit financing funds (Steering Group 1997), partly to dampen the negative effects of the market reforms. In spite of this, political and bureaucratic actors seized upon regional health authorities as the solution to winning support for the reforms. The minister of health wanted to use the regional health authorities as a contrast to the crown health enterprises to help sell the reforms to the public. The minister considered that the implementation of RHAs “should not be seen as fully commercial enterprises” and that there was “a responsibility for community consultation which in his mind [the minister of health] appeared to involve some degree of selling the health reforms at the community level” (Ministry of Health Files 1992h: n.p.). Advisers thought that regional health authorities would probably be perceived as unresponsive to local interests and concerns, as they were being established “when their focus is more likely to be on organization building and the initial contracting round” (ibid.). These discussions of the role of the market and the difficulties of reform played out in the media, shaping public opinion throughout the reform period. To understand this empirically, a data set containing news stories for the period February 1991 to August 1997 was created. The data set was based on a search of newspaper stories using the New Zealand Ministry of Health Library. The library clipped press stories on all aspects of health care and public health from community, city, and national papers in New Zealand. The major focus of the database was newspapers, although radio transcripts, cartoons, and press releases were also kept in the database. The Ministry of Health’s media database was searched for articles between February 1991 and August 1997 for material that contained the phrase health reform. Each story in the database listed the headline title of the news story, the newspaper it was published in, the date, the type of



1092


Table 8

Tone of Media Coverage of the Health Reforms, 1991–1997

Tone

1991 1992 1993 1994 1995 1996 1997 Total

Neutral implications for reform Negative implications for reform Positive implications for reform Total %

%

20

91

88

23

20

11

2

255 19%

48

230

287

181

103

60

11

920 68%

10

53

64

25

12

10

1

175 13%

78

374

439

229

135

81

14

6%

28% 33% 17% 10%

6%

1350

1%

Source: Health Reforms Media Database, compiled by the author.

clipping (letter, editorial, cartoon, or news), and the first sentence of the news story, or the lead. The search yielded 1,390 records. There were 1,134 news stories in the data set, 181 letters to the editor, and 35 editorials. Of these records, forty were excluded because they were cartoons that could not be coded or press releases. After removing the forty cartoons, the final data set had 1,350 records. Variables of interest were created, including the type of record, the year of publication, the location of the newspaper (using thirteen regional classifications), and the tone of reportage. Tone was coded using a code of neutral, positive, or negative. Table 8 shows that of the 1,350 records, 28 percent of the stories were published in 1992, when health reform was being considered, and 33 percent of the stories were published in 1993, the official year of implementation. Stories about health care reform declined over time, so that only 1 percent of the database stories were in 1997, although in 1997 the database only had records from the first eight months of the year. Around two-thirds (68 percent) of the reportage was coded as negative, 19 percent was neutral, and only 13 percent was positive. This database is large and it covers the major daily and weekly newspapers and provincial and rural papers. Furthermore, the inclusion of letters to the editor and editorials also allows analysis of public debate over the reforms. However, the limitation is that the full text of the stories is not accessible, since the records extracted from the Ministry of Health only contained the leader and headline of the article.



Laugesen

■


Lessons on Market Reforms in NHS Systems

In the first section of the essay, market reforms that were proposed and implemented in Italy, Spain, Sweden, and the United Kingdom were discussed, but not the fate of the reforms. The reforms in these national health service systems show some common patterns. As in New Zealand, after the implementation of reforms there was a counterreaction in most of these countries against competition as the major mechanism to increase efficiency and choice and simultaneously a solidification of agency-driven policies, considered first. The lessons from agency theory were more long lasting, perhaps because this approach is inherently flexible and able to be adapted by different political parties. All governments in the 1990s used the purchaserprovider split. The more rigid version of this fell out of favor, particularly in the United Kingdom. User fees have remained. Hospital management reforms essentially have remained in place, but again with some subtle modifications and shifts toward clinical rather than only managerial influence over hospitals. However, policies to encourage greater managerial autonomy, either from government or local voters, have been more long lasting, with almost all countries retaining this policy. In Italy, 1999 reforms revised the “cost paring mentality” and gave more power to doctors while limiting the powers on chief executive officers and strengthening local authority control (France and Taroni 2005). In Spain, regional governments implemented new public management reforms and experiments in private providers and new organizational instruments were part of a second wave of reforms (Rico and Costa-Font 2005). However, regional governments pursued these strategies simultaneously as they advocated expanding the scope of the national health service. Regional governments allied with trade unions and citizen groups in opposing “cost-minded central government politicians” (ibid.). Managed competition proposals were strongly opposed. In Sweden, the sale of one public hospital created a reaction against privatization, and a law introduced in 2000 banned such sales, while limits on contracting with the private sector were also recommended (Saltman and Bergman 2005). In general, Sweden’s approach was to engage in a careful calibration of the market by encouraging managerial independence and competition, but public funding has never been seriously questioned (ibid.). In the United Kingdom the overall ethos shifted from competition to collaborative relationships and finally to central “command and control” policies (Le Grand 1999, 2002) driven by the central government under a strong electoral mandate



1094


by the Blair government. Competition was minimized early on to avoid political embarrassment and preserve equity (Light 2001b). Furthermore, although reformers in many countries initially saw reforms as reducing the role of the state, shifting control to markets means more governmental effort has to be expended on enforcing quality, good value, and equity, and the government is required to be involved in setting standards and clinical guidelines and monitoring services (Light 2001a: 1159). The move away from the commercial model saw increasing centralization for decision making and greater bureaucratic control of the health care system. Britain’s central government asserted its authority through service frameworks, national treatment guidelines, and performance assessment (Le Grand 1999, 2002). The rhetoric around health care reform has thus shifted considerably. Trusts will become a model of “social ownership” modeled on cooperative societies and mutual organizations with trusts being governed by boards of governors elected by patients, the public, and employees and also accountable to various local organizations and national regulators (Klein 2003). No country with national health services described here has ventured into voucher-style financing. Financing reforms designed to introduce competition between funding sources for enrollees, or voucher-type arrangements, have been the least successful of all policies and were usually only proposed. The voucher-type financing plan was proposed in the United Kingdom and Spain, reversed in Italy, and New Zealand included an option in its legislation but never implemented this option. Is the shift toward a more public role in health care simply explained by standard variables such as the success of leftist parties and the economy later in the 1990s, or the institutions? After all, socialist or leftist victories occurred in the countries that allowed those changes to take place. However, the causality is not necessarily obvious—in many cases, parties were elected partly because of health care reform failures, especially in the United Kingdom, and governments changed because of a perception that reformers had ignored public preferences. In Spain, “a democratic deficit” also drove 1999 reforms (France and Taroni 2005). Economic conditions may have played a role, since all countries began the 1990s in fiscal distress, but experienced greater growth later in the decade, allowing more expansionary policies. This explanation would seem logical, but under this explanation we would expect to see policies rejected at the point of economic improvement, whereas they usually were adapted close to the time of adoption, when the economy was still performing poorly.



Laugesen

■


However, the economic conditions may have allowed a fuller shift to greater public funding than if the countries were still in recessions. Political institutions exert some influence. New Zealand’s more pure Westminster institutional structure was instrumental in enabling the cabinet to quickly and decisively shift the health system toward a more commercial model in New Zealand (Laugesen 2001). However, all five countries shifted toward more public funding and regulation in spite of different national legislative institutions. It was important that the larger constitutional structure of regionalism (in Italy and Spain), localism (in Sweden), and centralization (in New Zealand and the United Kingdom) did exert an influence on the reforms considered and implemented, but not on the fundamental public-private decisions. Conclusion

Five reform strategies were pursued in the 1990s by governments that favored market-oriented reforms: (1) increased cost sharing for patients through user fees, (2) the separation of purchaser-provider functions, (3) management reforms of hospitals, sometimes paired with profit or return on capital expectations, (4) provider competition, and (5) vouchers for purchasing health insurance. These policies were partly derived from agency theory and from a model of managed competition in health insurance. In most countries these policy reforms were proposed to solve problems of inefficiency, consumer responsiveness, and productivity. In Italy, New Zealand, Spain, Sweden, and the United Kingdom, some of these market reforms were adopted, while other policies were firmly rejected. The separation of purchaser and provider was typically more fluid than first envisaged, but nevertheless was enduring and important across all countries, even if the national government became the purchaser. Managerial reforms within hospitals occurred in all countries. Provider competition has been limited, with traces in Italy and Sweden, but competition is not the major instrument of policy makers. Plans for greater cost sharing were modified or dropped altogether. Public opposition stalled proposals to give people the option of choosing health insurance plans from competing social or private insurers. The market model was taken the furthest in New Zealand, but the analysis here shows that the model was short lived. Whereas elites thought that the Thatcher revolution would extend to health care, voters saw health care differently. Commercial aspects of New Zealand’s reforms were espe-



1096


cially disconcerting for voters. In addition, developing a market structure in the health sector is difficult for reasons of scale alone, but also because of the potential for public conflict with professionals, played out in the media. In the blame game, individual voters are very likely to take cues from sympathetic professional groups opposed to reform. The choice of market policies in the five countries suggests that some market reforms are more legitimate than others. Agency-focused reforms, purchaser-provider splits, and managerial reforms were adopted in all countries. The exception is agency reforms that shift risk to individuals. Voters do not welcome paying a higher proportion of their health care costs. Reforms that deal with agency problems by strengthening accountability and improving management of resources are the softer and more durable part of market reforms. Competition-based reforms, either at the provider or financing level, sit uncomfortably in a national health service model. The adoption of one type of reform over another may originate in the preexisting tendency toward hierarchical distribution of power over lateral distribution. However, there is evidence to suggest that the adoption of market competition as the major force for resource allocation in a health care system is less likely when a majority of people see health care as different from other parts of the economy.

References Alchain, Armen A., and Harold Demsetz. 1972. Production, Information Costs, and Economic Organization. American Economic Review 62: 777–795. Anell, A. 1996. The Monopolistic Integrated Model and Health Care Reform: The Swedish Experience. Health Policy 37: 19–33. Barnett, P., R. Perkins, and M. Powell. 2001. On a Hiding to Nothing? Assessing the Corporate Governance of Hospital and Health Services in New Zealand, 1993– 1998. International Journal of Health Planning and Management 16: 139–154. Bay of Plenty Times. 1992. Reforms Offer Faster Service. June 3. Bayle, M. S., and H. B. Cal. 2001. The People’s Campaign against Health Care Counter-Reforms in Spain. Journal of Public Health Policy 22: 139–152. Boston, Jonathan. 1991a. Reorganizing the Machinery of Government: Objectives and Outcomes. In Reshaping the State: New Zealand’s Bureaucratic Revolution, ed. J. Boston, J. Martin, J. Pallot, and P. Walsh, 233–267. Auckland: Oxford University Press. ———. 1991b. The Theoretical Underpinnings of Public Sector Restructuring in New Zealand. In Reshaping the State: New Zealand’s Bureaucratic Revolution, ed. J. Boston, J. Martin, J. Pallot, and P. Walsh, 1–26. Auckland: Oxford University Press.



Laugesen

■


Department of Prime Minister and Cabinet. 1992. Report and Recommendations to the Social Services Committee on the Submissions Received on the Health and Disability Services Bill. Wellington: Department of Prime Minister and Cabinet. Devlin, N., A. Maynard, and N. Mays. 2001. New Zealand’s New Health Sector Reforms: Back to the Future? British Medical Journal 322: 1171–1174. Donatini, Andrea, Ana Rico, Maria Giuseppina D’Ambrosio, Alessandra Lo Scalz, Letizia Orzella, Americo Cicchetti, and Silvia Profili. 2001. Italy: Health Care Systems in Transition. Copenhagen: European Observatory on Health Care Systems. English, B. 1998. Hospital Services Plan: Securing Better Hospital Services into the Future. Wellington: Ministry of Health. Enthoven, Alain C. 1988. Theory and Practice of Managed Competition in Health Care Finance. Vol. 9, Professor Dr. F. De Vries Lectures in Economics Theory, Institutions, Policy. Amsterdam: North Holland. Evans, Lewis, Arthur Grimes, Bryce Wilkinson, and David Teece. 1996. Economic Reform in New Zealand 1984–95: The Pursuit of Efficiency. Journal of Economic Literature 34: 1856–1902. Evening Post. 2000. Labour to Abolish Market-Led Health Model. August 2. Fougere, Geoff, Claudia Scott, and John Marwick. 1986. Report of the Health Benefits Review. Wellington: Government Printing Office. France, George, and Francesco Taroni. 2005. The Evolution of Health-Policy Making in Italy. Journal of Health Politics, Policy and Law 30: 169–187. Gaminde, I. 1999. Priorities in Healthcare: A Perspective from Spain. Health Policy 50: 55–70. Ham, C. 1996. Population-Centered and Patient-Focused Purchasing: The U.K. Experience. Milbank Quarterly 74: 191–214. Hospital and Related Services Taskforce. 1988. Unshackling the Hospitals. Wellington: Government Printing Office. Iyengar, Shanto. 1990. Shortcuts to Political Knowledge: The Role of Selective Attention and Accessibility. In Information and Democratic Processes, ed. J. A. Ferejohn and J. H. Kuklinski, 160–185. Urbana: University of Illinois Press. Jensen, Michael C., and William H. Meckling. 1976. Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure. Journal of Financial Economics 3: 305–360. Jérôme-Forget, Monique, Joseph White, and Joshua M. Wiener. 1994. Introduction. In Health Care Reform through Internal Markets: Experience and Proposals, ed. M. Jérôme-Forget, J. White, and J. M. Wiener, 7–16. Montreal: Institute for Research on Public Policy; Washington, DC: Brookings Institution. Kahneman, Daniel, and Amos Tversky. 1984. Choices, Values, and Frames. American Psychologist 39: 341–350. Khong, Yuen Foong. 1992. Analogies at War. Princeton, NJ: Princeton University Press. Klein, R. 2003. Governance for NHS Foundation Trusts. British Medical Journal 326: 174–175. Laugesen, M. J. 2001. The Institutional Context. In Health and Public Policy in New Zealand, ed. P. Davis and T. Ashton, 127–143. Auckland: Oxford University Press.



1098


Le Grand, J. 1999. Competition, Cooperation, or Control? Tales from the British National Health Service. Health Affairs 18(3): 27–39. ———. 2002. Further Tales from the British National Health Service. Health Affairs 21(3): 116–128. Light, D. W. 2001a. Comparative Institutional Response to Economic Policy Managed Competition and Governmentality. Social Science and Medicine 52: 1151–1166. ———. 2001b. Managed Competition, Governmentality, and Institutional Response in the United Kingdom. Social Science and Medicine 52: 1167–1181. Manukau Courier. 1992. More Say to Consumers. February 18. McGraw, Kathleen M. 1990. Avoiding Blame: An Experimental Investigation of Political Excuses and Justifications. British Journal of Political Science 20: 119–131. Ministry of Health. 1996. Sustainable Funding Package for the Health and Disability Sector. Wellington: Ministry of Health. ———. 1999. The Government’s Medium-Term Strategy for Health and Disability Support Services. Wellington: Ministry of Health. Ministry of Health Files. 1991a. Funding Options, ed. L. Hawkins, March 22. Wellington: Ministry of Health. ———. 1991b. Memo to Minister of Health from Chairman, Health Services Taskforce, May 10. Wellington: Ministry of Health. ———. 1991c. Memorandum for Cabinet. Reform of the Funding and Provision of Health Services, ed. L. Hawkins and C. Scott, June 7. Office of the Prime Minister. Wellington: Ministry of Health. ———. 1991d. Memorandum for the Prime Ministerial Committee on the Reform of Social Assistance, May 2. Wellington: Ministry of Health. ———. 1991e. Options for Social Insurance Financing of Health. Annex to Memorandum for the Prime Ministerial Committee on the Reform of Social Assistance, ed. C. Rod Carr, May 2. [Contributors to the paper derived from separate list of papers completed for the task force indicating authorship.] Wellington: Ministry of Health. ———. 1992a. Health Reforms Preliminary Timetable, ed. Department of Prime Minister and Cabinet to Minister of Health and Minister of Crown Health Enterprise. Wellington: Ministry of Health. ———. 1992b. Memo from Loraine Hawkins to David Smyth—Comment on Social Assistance Reforms Paper, ed. L. H. (Treasury), July 2. Wellington: Ministry of Health. ———. 1992c. Memo from Minister of Health to Chairman, Cabinet Social Assistance Reform Committee. Wellington: Ministry of Health. ———. 1992d. Memo to Paul Gini from Phil Barry, March 17. Wellington: Ministry of Health. ———. 1992e. Memo to David Smyth from Phil Barry, Re: Work Programs: Establishing RHAs, ed. P. Barry. Wellington: Ministry of Health. ———. 1992f. Memo to Minister of Finance from Loraine Hawkins, Re: Interim Targeting Regime, January 17. Wellington: Ministry of Health. ———. 1992g. Memo to User Charges Steering Committee from Paul Gini, Re: High Use Health Card Anomalies. Wellington: Ministry of Health.



Laugesen

■


———. 1992h. Notes from Meeting with the Minister of Health, December 30. Wellington: Ministry of Health. ———. 1992i. Social Assistance Reform Committee Cabinet Paper CAB 92 M 47/17. Wellington: Ministry of Health. ———. 1992j. Treasury Report to Minister of Finance from Loraine Hawkins, Re: Health User Charges: Long Term Strategy, February 27. Wellington: Ministry of Health. ———. 1992k. Treasury Report to Minister of Finance Health on User Charges: Long Term Strategy. Wellington: Ministry of Health. ———. 1993a. File Note. Post-Election Briefing CHEMU (Crown Health Enterprises Monitoring Unit), October 21. Wellington: Ministry of Health. ———. 1993b. Health System Survey Topline Results, ed. C.M. Research Associates. Prepared for the Health Services Information Office. Wellington: Ministry of Health. ———. 1993c. Memo from Lindsey Lawton, Health Sector Reform Management Unit, to David Smythe, Deputy Director General. Re: Market Development Survey: Draft Report, November 25. Wellington: Ministry of Health. ———. 1993d. Memo to Ian McPherson and Alf Kirk, Health Services Information Office, from Evan Voyce, Crown Health Enterprise Establishment Unit. Re: Communications Strategy for Health Reforms, May 20. Wellington: Ministry of Health. Nagel, Jack H. 1998. Social Choice in a Pluralitarian Democracy: The Politics of Market Liberalization in New Zealand. British Journal of Political Science 28: 223–267. National Archives. 1991. Minutes and Project Planning for the Review of Benefits 1991 and Interim Targeting. Memo from Steve Anderson, Department of Health, August 23. Wellington: National Archives. New Zealand Election Studies. 1993. www.nzes.org/exec/show/1993. ———. 1996. www.nzes.org/exec/show/1996. ———. 1999. www.nzes.org/exec/show/1999. New Zealand Treasury. 2005. Completed Asset Sales 2005. www.treasury.govt.nz/ assetsales/completed3.asp. ———. 2001. Health Data 2001 (CD Rom). Paris: OECD. Organisation for Economic Co-operation and Development. 1992. The Reform of Health Care: A Comparative Analysis of Seven OECD Countries. Paris: OECD. Pierson, Paul. 1994. Dismantling the Welfare State? New York: Cambridge University Press. Press (Christchurch). 1992. Minister Predicts More Flexible Health Services, February 3. Rico, Ana, and Joan Costa-Font. 2005. Power Rather than Path Dependency? The Dynamics of Institutional Change under Health Care Federalism. Journal of Health Politics, Policy and Law 30: 231–252. Rico, Ana, and Ramon Sabes. 2000. Spain: Health Care Systems in Transition. Copenhagen: European Observatory on Health Care Systems.



1100


Robinson, Ray, and Anna Dixon. 1999. United Kingdom: Health Care Systems in Transition. Copenhagen: European Observatory on Health Care Systems. Rodriguez, M., R. M. Scheffler, and J. D. Agnew. 2000. An Update on Spain’s Health Care System: Is It Time for Managed Competition? Health Policy 51: 109–131. Rosenthal, M., and M. Schlesinger. 2002. Not Afraid to Blame: The Neglected Role of Blame Attribution in Medical Consumerism and Some Implications for Health Policy. Milbank Quarterly 80: 41–95. Saltman, Richard B., and Sven-Eric Bergman. 2005. Renovating the Commons: Swedish Health Care Reforms in Perspective. Journal of Health Politics, Policy and Law 30: 253–275. Schlesinger, M., and T. K. Lee. 1993. Is Health Care Different? Popular Support of Federal Health and Social Policies. Journal of Health Politics, Policy and Law 18: 555–621. Steering Group. 1997. Implementing the Coalition Agreement: Report of the Steering Group to Oversee the Changes in the Health and Disability Sector. New Zealand Government: Wellington. Stone, Deborah A. 1989. Causal Stories and the Formation of Policy Agendas. Political Science Quarterly 104: 281–300. Upton, Simon. 1991. Your Health and the Public Health: A Statement of Government Policy by the Hon. Simon Upton, Minister of Health. Wellington: Minister of Health. Weaver, R. K. 1986. The Politics of Blame Avoidance. Journal of Public Policy 6: 371–398.
